X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FExtMsg.pm;h=549e44955e8c6db0810f622ea77722e189d77cf9;hb=11eec3365d537b20ade6923eaa5fdd18206d4342;hp=14d49cc59e186c8102fe60eda13513c767a2fa94;hpb=cf439203c31836e4c48f632f902aa58db6924972;p=public-inbox.git diff --git a/lib/PublicInbox/ExtMsg.pm b/lib/PublicInbox/ExtMsg.pm index 14d49cc5..549e4495 100644 --- a/lib/PublicInbox/ExtMsg.pm +++ b/lib/PublicInbox/ExtMsg.pm @@ -1,4 +1,4 @@ -# Copyright (C) 2015-2018 all contributors +# Copyright (C) 2015-2019 all contributors # License: AGPL-3.0+ # # Used by the web interface to link to messages outside of the our @@ -8,13 +8,13 @@ package PublicInbox::ExtMsg; use strict; use warnings; -use PublicInbox::Hval; +use PublicInbox::Hval qw(ascii_html prurl); use PublicInbox::MID qw/mid2path/; use PublicInbox::WwwStream; our $MIN_PARTIAL_LEN = 16; # TODO: user-configurable -our @EXT_URL = ( +our @EXT_URL = map { ascii_html($_) } ( # leading "//" denotes protocol-relative (http:// or https://) '//marc.info/?i=%s', '//www.mail-archive.com/search?l=mid&q=%s', @@ -29,6 +29,10 @@ our @EXT_URL = ( sub PARTIAL_MAX () { 100 } +sub mids_from_mset { # Search::retry_reopen callback + [ map { PublicInbox::SearchMsg::from_mitem($_)->mid } $_[0]->items ]; +} + sub search_partial ($$) { my ($srch, $mid) = @_; return if length($mid) < $MIN_PARTIAL_LEN; @@ -65,12 +69,28 @@ sub search_partial ($$) { # Search::Xapian::QueryParserError or even: # "something terrible happened at ../Search/Xapian/Enquire.pm" my $mset = eval { $srch->query($m, $opt) } or next; + my $mids = $srch->retry_reopen(\&mids_from_mset, $mset); + return $mids if scalar(@$mids); + } +} + +sub ext_msg_i { + my ($other, $arg) = @_; + my ($cur, $mid, $ibxs, $found) = @$arg; - my @mids = map { - my $doc = $_->get_document; - PublicInbox::SearchMsg->load_doc($doc)->mid; - } $mset->items; - return \@mids if scalar(@mids); + return if $other->{name} eq $cur->{name} || !$other->base_url; + + my $mm = $other->mm or return; + + # try to find the URL with Msgmap to avoid forking + my $num = $mm->num_for($mid); + if (defined $num) { + push @$found, $other; + } else { + # no point in trying the fork fallback if we + # know Xapian is up-to-date but missing the + # message in the current repo + push @$ibxs, $other; } } @@ -80,28 +100,13 @@ sub ext_msg { my $mid = $ctx->{mid}; eval { require PublicInbox::Msgmap }; - my $have_mm = $@ ? 0 : 1; - my (@ibx, @found); - - $ctx->{www}->{pi_config}->each_inbox(sub { - my ($other) = @_; - return if $other->{name} eq $cur->{name} || !$other->base_url; - - my $mm = $other->mm or return; - - # try to find the URL with Msgmap to avoid forking - my $num = $mm->num_for($mid); - if (defined $num) { - push @found, $other; - } else { - # no point in trying the fork fallback if we - # know Xapian is up-to-date but missing the - # message in the current repo - push @ibx, $other; - } - }); + my $ibxs = []; + my $found = []; + my $arg = [ $cur, $mid, $ibxs, $found ]; + + $ctx->{www}->{pi_config}->each_inbox(\&ext_msg_i, $arg); - return exact($ctx, \@found, $mid) if @found; + return exact($ctx, $found, $mid) if @$found; # fall back to partial MID matching my @partial; @@ -115,7 +120,7 @@ sub ext_msg { # can't find a partial match in current inbox, try the others: if (!$n_partial && length($mid) >= $MIN_PARTIAL_LEN) { - foreach my $ibx (@ibx) { + foreach my $ibx (@$ibxs) { $srch = $ibx->search or next; $mids = search_partial($srch, $mid) or next; $n_partial += scalar(@$mids); @@ -167,7 +172,7 @@ sub ext_urls { my $env = $ctx->{env}; my $e = "\nPerhaps try an external site:\n\n"; foreach my $url (@EXT_URL) { - my $u = PublicInbox::Hval::prurl($env, $url); + my $u = prurl($env, $url); my $r = sprintf($u, $href); my $t = sprintf($u, $html); $e .= qq{$t\n};